July 2, 1894.] 



KNOWLEDGE. 



163 



hottest months. Unfortunately, the same cannot be said 

 of the bite of the Indian and African ^dpers, and of the 

 various species of American rattlesnakes. 



TYPES OF FLORAL STRUCTURE. 



By the Rev. Alex. S. Wilson, M.A., B.Sc. 



A FLOWER commonly consists of organs of four 

 different kinds, arranged in concentric circles or 

 whorls. The order in which these organs occur 

 is always the same relatively to the centre of the 

 flower. The carpels forming the gynrecium are 

 most central ; nest come the stamens composing the 

 andrcecium ; outside of these is the corolla, made up of 

 petals, invested externally by the sepals, which collectively 

 constitute the calyx. There is no apparent reason why 

 the carpels and stamens should not occasionally change 

 places, and an explanation of this invariable order is still a 

 ilcsideratwn. 



A second very obvious character of flowers, which admits 

 of better explanation, is the prevalence of certain numbers 

 among the members composing the different whorls. 

 There is a constant recurrence of the same number of 

 parts in the different whorls of the same flower, and in 

 corresponding whorls of different flowers. The number 

 five is exceedingly common. Thus in the violet, with its 

 five sepals, five petals, and five stamens, we have an 

 example of a pentamerous flower. Three is another 

 favourite number, and of this ternary symmetry the iris, 

 with its three sepals, three petals, three stamens, and three 

 carpels, is a good instance. 



By founding his classification of plants so largely upon 

 the number ol the floral organs, Linnteus gave prominence 

 to this numerical character, and, highly artificial though 

 it be, the Linufean system has the merit of recognizing 

 the fundamental importance of this feature as an index of 

 natural affinity. 



When we study phyllotaxis, or the modes in which 

 leaves are arranged on stems and branches, the reason for 

 the prevalence of certain numbers in flowers becomes 

 apparent. Leaves on a plant's stem are not arranged at 

 random, but according to a definite law. Avery frequent 

 arrangement is the alternate, where they are separated 

 by equal spaces or internodes, and so placed that a line 

 drawn through the bases of successive leaves describes an 

 ascending spiral. This line, winding round the stem, is 

 known as the ijeneratiwi spiral ; the fraction of the stem's 

 circumference which it traverses in passing from one leaf 

 to the next is the anfjidar ilireriience, and the Icaf-ryde is 

 the portion of the generating spiral included between one 

 leaf and the next above it in the same vertical line. 

 Angular divergence is expressed by such fractions as 

 21 ii f> f • ^s> -ii' ^^-t wbere the denomLnator indicates 

 the number of leaves encountered, and the numerator the 

 number of revolutions round the stem made, by the 

 generating spiral in completing the cycle — that is, in 

 passing from one leaf to the next above it in the same 

 perpendicular line. 



Frequently, though not always, the angular divergence 

 is the same for the stem and branches of the same plant ; 

 it is generally constant for the same species, and may even 

 be characteristic of genera and natural orders. A common 

 case is where the leaves are separated from each other by 

 one third of the circumference of the stem ; the angular 

 divergence in this case is ^ and the cycle includes three 

 leaves. A still more frequent arrangement gives an 

 angular divergence of § with five leaves and two turns of 

 the generating spiral to the cycle. Many plants, again, 



produce their leaves in pairs, one on each side of the stem ; 

 since the vertical planes of the successive pairs commonly 

 cut each other at right angles, this arrangement, well 

 exemplified in the mint and other Labiatae, is described 

 as opposite and decussate. Three, four or more leaves 

 springing from the same node constitute a whorl or 

 verticil ; such a verticillate arrangement is seen in the 

 leaves of the wood-ruff' and other Rubiacese. The whorling 

 of leaves may be regarded as resulting from the suppression 

 of internodes at regular intervals. A shoot having three 

 leaves placed according to the \ arrangement would, if we 

 suppose its internodes to remam undeveloped, give rise to 

 a whorl of three leaves ; similarly a shoot with an angular 

 divergence of f would furnish whorls composed of five 

 leaves if its internodes were suppressed. Hence it is held 

 that each flower whorl represents a compressed leaf-cycle, 

 the number of its constituent parts depending on the leaf 

 arrangement to which it corresponds. The frequency of 

 the numbers 3 and 5 among the organs of the flower 

 becomes intelligible in view of the prevalence among 

 foliage leaves of the J and | arrangements. 



Between phyllotaxis or leaf arrangement and floral 

 symmetry there is thus a very close connection, and we 

 might therefore be led to expect that they would always 

 be in agreement ; we might not unnaturally suppose, for 

 example, that every plant with pentamerous flowers 

 should have its leaves arranged on the ?- plan and vice versa. 

 Although this rule holds good for many plants, it is very 

 far from universal, and the numerous exceptions point to 

 the conclusion that floral organs have more frequently 

 retained the ancestral phyllotaxis than foliage leaves, 

 which in many cases seem to have departed widely from 

 the primitive arrangement. In other words, the hereditary 

 tendency asserts itself much more strongly in the arrange- 

 ment of the floral organs than in that of foliage ; hence 

 the importance assigned to tlie flower in every system of 

 classification. 



From cases of reversion like the flowering cherry, where 

 the carpels are replaced by ordinary green leaves, we learn 

 that a flower is simply an arrested branch or leafy shoot 

 having its internodes undeveloped and its leaves modified 

 to subserve the function of reproduction. The passage 

 from vascular cryptogams to phanerogams mvolved, as 

 was shown in a previous article (Vol. XVII., p. 125), the 

 arrest of certain structures belonging to the seed and 

 pollen grain ; it now appears that in the formation of the 

 flower we have a further illustration of arrested develop- 

 ment. 



Excluding the Coniferae and their allies, which are 

 gymnospermous, aO flowering plants have their seeds 

 protected by closed carpels, and are on this account 

 designated angiosperms. Plants embraced under this 

 designation are either monocotyledons or dicotyledons, as 

 the two great classes are called into which angiosperms 

 are divided. To which of these classes a plant belongs 

 can easily be ascertained from its flowers. Each whorl in 

 the flower of a monocotyledon consists of three parts, or of 

 a number which is a multiple of three. The flower of a 

 dicotyledon, on the other hand, is made up of whorls each 

 commonly composed of five parts, or a multiple of five ; 

 less frequently, as in the fuchsia and waU-flower, of four, 

 or a mifltiple of four. Besides this distinction, there is a 

 difl'erence in their floral envelopes ; a monocotyledon has 

 the sepals and petals alike — thus in the tulip both whorls 

 of the perianth are coloured or petaloid. Dicotyledons, 

 agam, have the sepals mostly green, the petals alone 

 being gaily coloured. The two classes are farther 

 distinguished by the characters of their vegetative organs ; 

 monocotyledons have parallel-veined leaves, endogenous 



